Chapter 2: Module

EARTHFILL 4- Embankment
DAMS dams
Earthfill dams
Earthfill dams are the most common type of dam built to any height. They are
designed as a non-overflow section with separate spillway. The reason for such
wide spread use of earthfill dams are:
• The foundation requirements are not as rigorous as other dams
• Local available soil is the main construction material
• High skill not required
• No special plants are required. Most earth-moving machines can be used
•

Fig. 1. General view of an Earthfill Figure 2: General view of an Earthfill

dam dam

Classification of earthfill dams

Earthfill dams are classified by many factors.

1. Based on the method of construction

• Rolled fill earth dams
• Hydraulic fill dam

2. Based on mechanical characteristics of earth materials making the section of

the Dam
• Homogeneous earth dams
• Non-Homogeneous (zoned) earth dams
i) non-homogeneous with inclined impervious zone (ekran) of
artificial material
ii) with impervious zone of soil with low permeability
iii) with central core soil material of low permeability
iv) with a central thin diaphragm of impervious material
Rolled Fill Earth dams

5
In this type of dams, successive layers of moistened or damp soils are laid one
over the other. Each layer not exceeding 20 cm in thickness is properly
consolidated at optimum moisture content, only then is the next layer laid.

Hydraulic Fill dams

In this type of dams, the construction, excavation, transportation of the earth is
done by hydraulic methods. Outer edges of the embankments are kept slightly
higher than the middle portion of each layer. During construction, a mixture of
excavated materials in slurry condition is pumped and discharged at the edges.
This slurry of excavated materials and water consists of coarse and fine
materials. When it is discharged near the outer edges, the coarser materials
settle first at the edges, while the finer materials move to the middle and settle
there. Fine particles are deposited in the central portion to form a water tight
central core. In this method, compaction is not required.

Homogeneous Earthen Dam

These dams are constructed with uniform and homogeneous materials. It is
suitable for low height dams (up to 10m). These dams are usually constructed
with soil and grit mixed in proper ratios. The seepage action of such dams are
not favourable, therefore, for safety in case of rapid drawdown, the upstream
slope is kept relatively flat (3:1)
Homogeneous section is modified by constructing rock toe at the downstream
lower end and providing horizontal filter drain.

6
Zoned Earth Dams
These are dams with the central portions called core or hearting made from
materials which are relatively impervious. The thickness of the core wall is made
sufficiently thick to prevent leakage of water through the body of the dam.

Dam with a Diaphragm

This type of dam is constructed with pervious materials, with a thin impervious
diaphragm in the central part to prevent seepage of water. The thin impervious
diaphragm may be made of impervious clayey soil, cement concrete or masonry
or any impervious material The diaphragm can be constructed in the central
portion or on the upstream face of the dam. The main difference in zoned and
diaphragm type of dams depend on the thickness of the impervious core or
diaphragm. The thickness of the diaphragm is not more than 10 m.

7
The criteria for the design of earth dams are:
1. Sufficient spillway capacity and freeboard are provided so that there is no
danger of overtopping of the dam
2. Seepage flow through the embankment is controlled so that the amount lost
does not interfere with the objective of the dam and there is no erosion or
sloughing of soil. In this respect, seepage line should remain well within the
downstream face of the dam and the portion of the dam on downstream side of
the impervious core should be well drained.
3. Uplift pressure due to the seepage underneath is not enough to cause piping
4. The slopes of the embankment are stable under all conditions of reservoir
operation, including rapid drawdown and during steady seepage under full
reservoir.
5. The stresses imposed by the embankment upon the foundation are less than
the strength of material in the foundation with a suitable factor of safety.
6. The upstream face is properly protected ((stone pitching, riprap, revetment)
against erosion caused by wave action, and the downstream face is protected
(counter-booms, turfs) against the action of rain

Embankment Materials

a. Earth-fill materials.
(1) While most soils can be used for earth-fill construction as long as they are
insoluble and substantially inorganic, typical rock flours and clays with liquid
limits above 80 should generally be avoided. The term “soil” as used herein
includes such materials as soft sandstone or other rocks that break down into soil
during handling and compaction.

(2) If a fine-grained soil can be brought readily within the range of water contents
suitable for compaction and for operation of construction equipment, it can be
used for embankment construction. Some slow-drying impervious soils may be
unusable as embankment fill because of excessive moisture, and the reduction
of moisture content would be impracticable in some climatic areas because of
anticipated rainfall during construction. In other cases, soils may require
additional water to approach optimum water content for compaction. Even
ponding or sprinkling in borrow areas may be necessary. The use of fine-grained
soils having high water contents may cause high porewater pressures to develop
in the embankment under its own weight. Moisture penetration into dry hard
borrow material can be aided by ripping or plowing prior to sprinkling or ponding
operations.

(3) As it is generally difficult to reduce substantially the water content of

impervious soils, borrow areas containing impervious soils more than about 2 to
5 percent wet of optimum water content (depending upon their plasticity

8
characteristics) may be difficult to use in an embankment. However, this depends
upon local climatic conditions and the size and layout of the work, and must be
assessed for each project on an individual basis. The cost of using drier material
requiring a longer haul should be compared with the cost of using wetter
materials and flatter slopes. Other factors being equal, and if a choice is possible,
soils having a wide range of grain sizes (well-graded) are preferable to soils
having relatively uniform particle sizes, since the former usually are stronger, less
susceptible to piping, erosion, and liquefaction, and less compressible. Cobbles
and boulders in soils may add to the cost of construction since stone with
maximum dimensions greater than the thickness of the compacted layer must be
removed to permit proper compaction. Embankment soils that undergo
considerable shrinkage upon drying should be protected by adequate
thicknesses of non-shrinking fine-grained soils to reduce evaporation. Clay soils
should not be used as backfill in contact with concrete or masonry structures,
except in the impervious zone of an embankment.

(4) Most earth materials suitable for the impervious zone of an earth dam are
also suitable for the impervious zone of a rock-fill dam. When water loss must be
kept to a minimum (i.e., when the reservoir is used for long-term storage), and
fine-grained material is in short supply, resulting in a thin zone, the material used
in the core should have a low permeability. Where seepage loss is less
important, as in some flood control dams not used for storage, less impervious
material may be used in the impervious zone.

Some causes of failure of earthfill dams

Like most of engineering structures, earth dams may fail due to faulty design,
improper construction and poor maintenance practices, etc
The various causes of failure may be classified as:
a) Hydraulic failure
b) Seepage failure
c) Structural failure

a) Hydraulic failure:
Hydraulic accounts for over 40% of earth dam failure and may be due to one or
more of the following:

9
.
F.W.L

Body of dam after

failure

Fig. Dam failure by overtopping

i) By overtopping: When free board of dam or capacity of spillway is insufficient,

the flood water will pass over the dam and wash it downstream.
ii). Erosion of downstream toe: The toe of the dam at the downstream side may
be eroded due to i) heavy cross-current from spillway buckets, or ii) tail water.
When the toe of downstream is eroded, it will lead to failure of dam. This can be
prevented by providing a downstream slope pitching or a riprap up to a height
above the tail water depth Also, the side wall of the spillway should have
sufficient height and length to prevent possibility of cross flow towards the earth
embankment.
iii) Erosion of upstream surface: During winds, the waves developed near the
top water surface may cut into the soil of upstream dam face which may cause
slip of the upstream surface leading to failure. For preventing against such
failure, the upstream face should be protected with stone pitching or riprap
iv). Erosion of downstream face by gully formation: During heavy rains, the
flowing rain water over the downstream face can erode the surface, creating
gullies, which could lead to failure. To prevent such failures, the dam surface
should be properly maintained; all cuts filled on time and surface well grassed.
Berms could be provided at suitable heights and surface well drained.

10
Figure 3. Beginning of downstream failure

b). Seepage failure:

Seepage always occurs in the dams. If the magnitude is within design limits, it
may not harm the stability of the dam. However, if seepage is concentrated or
uncontrolled beyond limits, it will lead to failure of the dam. Following are some of
the various types of seepage failure.
i) Piping through dam body. When seepage starts through poor soils in the body
of the dam, small channels are formed which transport material downstream. As
more materials are transported downstream, the channels glow bigger and
bigger which could lead to wash out of dam

.
H.F.L
.

Channel formation

Failure of dam due to piping through dam

body

11
ii) Piping through foundation: When highly permeable cavities or fissures or
strata of gravel or coarse sand are present in the dam foundation, it may lead to
heavy seepage. The concentrated seepage at high rate will erode soil which will
cause increase flow of water and soil. As a result, the dam will settle or sink
leading to failure.

Failure due to piping

piping

iii) Sloughing of downstream side of dam:

The process of failure due to sloughing starts when the downstream toe of the
dam becomes saturated and starts getting eroded, causing small slump or slide
of the dam. The small slide leaves a relative steep face, which also becomes
saturated due to seepage and also slumps again and forms more unstable
surface. The process of saturation and slumping continues, leading to failure of
dam.

c) Structural Failure:
About 25% of failure is attributed to structural failure, which is mainly due to
shear failure causing slide along the slopes. The failure may be due to:
i) Slide in embankment: When the slopes of the embankments are too steep, the
embankment may slide causing failure. This might happen when there is a
sudden drawdown, which is critical for the upstream side because of the
development of extremely high pore pressures, which decreases the shearing
strength of the soil. The downstream side can also slide especially when dam is
full. Upstream embankment failure is not as serious as downstream failure.
ii) Foundation slide: When the foundation of an earthfill dam is composed of fine
silt, clay, or similar soft soil, the whole dam may slide due to water thrust. If
seams of fissured rocks, such as soft clay, or shale exist below the foundation,
the side thrust of the water pressure may shear the whole dam and cause its

12
failure. In such failure the top of the dam gets cracked and subsides, the lower
slopes moves outward and forms large mud waves near the dam heel.
iii) Faulty construction and poor maintenance: When during construction, the
compaction of the embankment is not properly done, it may lead to failure.
iv) Earthquake may cause the following types of failure to earthfill dams;
1 cracks may develop in the core wall, causing leakages and piping failure.
2. slow waves may set up due to shaking of reservoir bottom, and dam may
fail due to overtopping
3. settlement of dam which may reduce freeboard causing failure by
overtopping
4. slidding of natural hills causing damage to dam and its appurtenant
structures
5. Fault movement in the dam site reducing reservoir capacity and causing
overtopping.
6. Shear slide of dam
7. The sand below foundation may liquefy
8. Failure of slope pitching.

Some elements of earthfill dams

1.Crest of dam:
The crest width of dams should be sufficient to keep the seepage line within the
dam, when the reservoir is full. The crest width of the dam if road is not
envisaged should not be less than 3m for low and medium head dams and 6m
for high head dams. If road is envisaged, then the width of the dam is determined
according to the class of road and determined by the road code.
Top width could be determined by the following recommended formulae:
a) For very low dams top width is given by

B = H/5 + 3

b) For dams lower than 30m

B = 0.55(H)1/2 + H/5

For dams higher than 30m,

13
B = 1.65(H + 1.5)1/3

B = 1.67(H)1/2

Balustrades are provided at the end of the roads to prevent car falling off the
slopes.
2. Side slopes of dam: side slope of dams must satisfy the static stability.
However, since the stability computations can be done only after defining the
profile of the dam and determining the seepage line, it becomes necessary to
give an initial side slopes. Initial slope could be taken from the tables below.
(Taken from “hydraulic structures” –N. P. Rosanova)
Slope Material Side slopes depending on height of dam
of dam Less than 5m From 5m - From 10-15m
10m
Upstream Clayey 2 2.5 3
Sandy 2.5 – 2 3 – 2.5 3
Downstream Clayey 1.5 1.75 1.75
With filter Sandy 2 2 2
Downstream Clayey 1.75 2 2.25
without filter Sandy 2 2.25 2.25

In low head dams, usually one and constant side slope is used, however in
medium and high head dams, different side slopes are usually adopted to reduce
the volume of the dam.

Side Slope According to the recommendations of Terzaghi

No. Type of material Upstream Downstream

slope slope (H:V)
(H :V)
1 Well graded homogeneous soil 2.5:1 2:1
2 Homogeneous coarse silt 3:1 2.5:1
3 Homogeneous silt clay:
i) for dam height less than 15m 2.5:1 2:1
ii) for dam height more than 3:1 2.5:1
15m
4 Sand or sand and gravel with 3:1 2.5:1
clay core
5 Sand or sand and gravel with 2.5:1 2:1
reinforced concrete core wall

3. Berms: Berms are constructed at both the upstream and downstream side of
the dam for the purposes of observing the conditions of protections at the
slopes and their repairs and also for increasing the width of dam at the base

14
 with aim of increasing seepage length. It is also done when constructions
coffer dams are made part of the body of the dam. At the downstream side,
berms are done at an interval of 10 – 15m high. Width of berm is taken
between 1 – 2 m.

4. Free Board

Normal freeboard is the vertical distance between the normal pool level and the
crest of the dam. Minimum freeboard is the vertical distance between the high
flood level and the crest of dam.
The minimum height of freeboard is taken as 1.5 hw where hw is given by:
hw = 0.032 (V.F)1/2 + 0.763 – 0.271(F)1/4 for F, 32 km -----------(X)
and hw = 0.032 (V.F)1/2 for F . 32km -------------------------------------(Y)
where hw = wave height ( height of water from top to trough of waves in meters)
V = velocity of wind in km/hr
F = fetch or straight length of water expanse in km.
Free board values as recommended by U.S.B.R are given in table below.

Free Board by USBR

Spillway Type Dam Height in m Minimum freeboard

over M.W.L
Free spillway Any height 2 m to 3 m
Controlled Up to 60 m 2.5 m above top of
spillway gate s
Controlled More than 60 m 3.0 m above top of
spillway gates

5. Slope Protection (Revetment)

15
Upstream side protection: For protecting the upstream slope from deterioration
and damage from wave action, the slope is covered with different protective
material.
Rock riprap, either dry dumped stone boulders or hand-packed stone boulders
can be made. Stone pitching provided at slope of 1.5 : 1 to 2 : 1 for ordinary soil
material of dam and 3 : 1 for poor soil material. The revetment stones are fixed
at the toe of the dam to prevent slipping of the embankment. The thickness of the

Figure 3. Upstream protection with hand- Figure 4. Upstream protection

layed rip-rap

stone pitching is usually more than 60 cm. In most cases the stone pitching is
placed over gravel then sand cushion. Big sizes stones with their broader face
downwards are packed with each other by means of hammer.
Concrete, reinforced concrete slabs, steel plates, bituminous material pavement,
brick tile pavement could also be used. However, extensive survey done by US
Corps of Engineers in the 1940s o over 100 dams show that dry dumped riprap
stone pitching has the best efficiency in terms of failure rate.

16
Downstream protection
One of the cheapest, simple and effective methods of protecting the downstream
from rainfall and wind action is by planting green grass (turfs) on the slope.
Counter-boom could also be done.

17
 Figure 4. Downstream protection

6. Drainage: Drainage in earth dams is meant for lowering the seepage curve;
prevent seepage water from flowing onto the downstream slope, and
conveying seepage water through the body of the dam to the downstream
part of the dam. By its function, dam drainage must have two parts; an intake
structure (drainage trench) that allows seepage water from the body and
foundation of the dam, while at the same time preventing deformation due to
seepage and conveyance structure that transports the seepage water from
the dam. However, in most drainage, it is difficult to see clearly these two
parts.

18
 Hydrotechnical construction practice has work out many drainage systems
depending on the type of dam, materials of the foundation and body of dam.
Among some of the most commonly used drainage systems are:
i) Drainage prism: with many positive sides (advantages) but requires the use of
large quantity of stones
ii) A type of drainage prism in which the filter material of the drainage system is
laid to extend to a certain height on the downstream side. Such drainage
system is used when there could be rise of the tail water above the crest of
the prism.
iii) Flat horizontal drainage: It requires much smaller quantity of stones and
simplifies construction. It has the advantage of draining both the foundation
and body of dam and it is used mainly when the foundation is made up of
saturated material
iv) Combination of horizontal drain with the prism.
v) Horizontal piped drainage: consists of a pipe (tube), laid parallel to base of
the slope of dam.
vi) Horizontal stone drainage: a type of horizontal piped drain in which instead of
the pipe, a stone prism is used.

19
 Fig.. Diaphragm Earthen Dam. (Source: Michael and Ojha, 2012)
Design Criteria
Following main design criteria may be laid down for the safety of an earth dam:
1. To prevent hydraulic failures the dam must be so designed that erosion of the embankment is
prevented. For this purpose, the following steps should be followed:
(a) Spillway capacity is sufficient to pass the peak flow.
(b) Overtopping by wave action at maximum water level is prevented.
(c) The original height of structure is sufficient to maintain the minimum safe freeboard after
settlement has occurred.
(d) Erosion of the embankment due to wave action and surface runoff does not occur.
(e) The crest should be wide enough to withstand wave action and earthquake shock.
2. To prevent the failures due to seepage:
(a) Quantity of seepage water through the dam section and foundation should be limited.
(b) The seepage line should be well within the downstream face of the dam to prevent sloughing.
(c) Seepage water through the dam or foundation should not remove any particle or in other words
cause piping.
(d) There should not be any leakage of water from the upstream to the downstream face. Such leakage
may occur through conduits, at joints between earth and concrete sections or through holes made by
aquatic animals.
3. To prevent structural failures:
(a) The upstream and downstream slopes of the embankment should be stable under all loading
conditions to which they may be subjected including earthquake.
(b) The foundation shear stresses should be within the permissible limits of shear strength of the material.
Design of Earthen Dam
The preliminary design of earthen dam is done on the basis of past experiences. For designing purpose
several parameters, given below should be considered.
1. Top Width
2. Free Board
3. Settlement Allowance
4. Casing or Outer Shell
5. Cut-off Trench
6. Downstream Drainage System

1. Top Width: Minimum top width (W) should be such that it can enhance the practicability and protect
it against the wave action and earth wave shocks. Sometimes it is also used for transportation purposes. It
depends upon the height of the earthen dam and can be calculated as follows:

19
where H = the height of the dam (m), for Indian conditions it should not be less than 6 m.
Free board: It is the vertical distance between the top of the dam and the full supply level of the reservoir
or the added height. It acts as a safety measure for the dam against high flow condition that is waves and
runoff from storms greater than the design frequency from overtopping the embankment. The
Recommended values of free board for different heights of earthen dams, given by U.S.B.R., are given in
Table.
Table . Recommended Values of Free Board given by U.S.B.R.
Nature of spillway Height of dam Free board

Free Any Minimum 2 m and maximum 3 m over the maximum flood level

Controlled < 60 m 2.5 m above the top of the gate

Controlled > 60 m 3 m above the top of the gate

If fetch length or exposure is given then the free board can also be calculated by Hawksley's formula:

where, = wave height (m); Dm = fetch or exposure (m).

2. Settlement Allowance: It is the result of the settlement of the fill and foundation material resulting in
the decrease of dam storage. It depends upon the type of fill material and the method and speed of
construction. It varies from 10% of design height for hand compacted to 5% for machine compacted
earthfill.
3. Casing or Outer Shell: Its main function is to provide stability and protection to the core. Depending
upon the upstream and downstream slopes, a recommendation for the casing and outer shell slopes for
different types of soils given by Terzaghi is presented in Table 1.

Table. Recommended Slopes of Earthen Dam (Sources: S.K. Garg, 2008)

20
Cutoff Trench: It is provided to reduce the seepage through the foundation and also to reduce the piping
in the dam. It should be aligned in a way that its central line should be within the upstream face of the
impervious core. Its depth should be more than 1 m. Bottom width of cutoff trench (B) is calculated as:

where h = reservoir head above the ground surface (m); and d = depth of cutoff trench below the ground
surface (m).
4. Downstream Drainage System: It is performed by providing the filter material in the earthen dam
which is more pervious than the rest of the fill material. It reduces the pore water pressure thus adding
stability to the dam.
Three types of drains used for this purpose are:
a) Toe Drains
b) Horizontal Blanket
c) Chimney Drains.

Determination of parametric line by Casagrande’s method

Phreatic Line in Earth Dam

Phreatic line is also known as seepage line or saturation line. It is defined as an imaginary line within a
dam section, below which there is a positive hydrostatic pressure and above it there is a negative
hydrostatic pressure. The hydrostatic pressure represents atmospheric pressure which is equal to zero on
the face of phreatic line. Above the phreatic line, there is capillary zone, also called as capillary fringe, in
which the hydrostatic pressure is negative. The flow of seepage water, below the phreatic line, reduces the
effective weight of the soil; as a result shear strength of a soil is reduced due to increased intergranular
pressure in earth fill material.

1. Derivation of Phreatic Line with Filter

In this case, before going directly for derivation, the important features of phreatic line must be known.
From the experimental evidence, it has been found that, the seepage line is pushed down by the toe filter

21
and is very close to parabolic shape except at the junction point of the upstream face. The upstream face
of the dam represents 100% equipotential line when it is covered by the water; under this condition the
seepage line should be drawn perpendicular to this face at the junction point.
Casagrande method is used for deriving the phreatic line (Fig); the procedure is described as follows:

Fig.. Derivation of phreatic line in earth dam. (Source: Suresh, 2002)

a) Let the phreatic line is assumed to be a base parabola with its focus at point F, i.e. at the starting point
of the filter, FE.
b) AB is the upstream face of the earth dam and L is the horizontal projection of face AB on the water
surface. Measure the distance BC equal to 0.3L. Count point C as a starting point of base of parabola.
c) For deciding the position of directrix of the parabola, the principle which states that every point of
the parabola is at equidistance from the focus as well as directrix. Hence, considering C as a center
and CF as radius, an arc is drawn which cut the horizontal line CB at point O. Since CO= CF, hence
vertical line OH will be directrix of parabola.
d) The last point G of the parabola will fall at the middle of the points F and H.
e) The intermediate points of parabola are located on the principle that their distance from the focus and
directrix are the same. Here to locate the point P as an intermediate point, a vertical line DP is drawn at
any distance x from the F. Now considering the distance DH as radius with F as a center, an arc is drawn
which cuts the vertical DP at point P.
f) Now all there obtained points are joined by free hand to get the base parabola. However, this needs to
be corrected at the entry point, for the feature that phreatic line must be started from the point B only, not
from C. It should be sketched perpendicular to the upstream face AB, as it is 100 percent equipotential
line. Now phreatic line is sketched by free hand in such way that, it should be perpendicular to
face AB and meets to rest of the points of the parabola tangentially. In addition, the base parabola should
also be met perpendicular, to the downstream face of the dam at pointG.
13.3 Equation of Parabola
The equation of base parabola can be derived from its basic properties i.e. the distance of any point P(x, y)
on the parabola from its focus is the same as the distance of the point P(x, y) from directrix.
Thus we have,

22
Where, s = focal distance (FH)
From equation,

1
This is the desired equation of base parabola.
For deriving the expression of discharge (q) for the earth dam equipped with horizontal filter, the Darcy's
law is used. According to which, the discharge (q) through vertical section PD, is equal to:

2
Partial differentiation of Eqn.2, resulted

3
Substituting the value of in Eqn. 3, the rate of seepage flow through the dam is given by:

This is the expression for computing the rate of seepage discharge through the body of earthen dam, in
terms of focal distance s. The distance s can be determined either graphically or analytically.
Considering C as co-ordinate, the value of s can be obtained as:
From Eqn: 1

At point C, x = D and y = H

23
By using this equation, if the value of coefficient of permeability (k) and focal distance (s) are known, the
discharge (q) can be calculated. This gives an accurate value of seepage rate and is applicable to such
dams, which are provided with horizontal drainage (filter) system but can also be used for other types of
dam section.

2. Phreatic Line in Earthen Dam without Filter

The position of phreatic line in an earth dam without filter can be determined using the same manner, as
in previous case i.e. with a filter. In this case, the focal point (F) of the parabola will be the lowest point
of the downstream slope (Fig. 2). The base of the parabola BJC cut at a point Jon downstream slope and
is extended beyond the limit of the dam, as indicated by dotted line, but the seepage line should be
emerged at point K, tangential to downstream face. In this way, the phreatic line should be shifted to the
point K from J. The distance KF is known as discharge face, which always remains under saturation
condition. The correction JK (say) by which the base of parabola need to be shifted downward, can be
determined by graphical and analytical methods.

Fig. 2. Phreatic line without filter.

(Source: Suresh, 2002)
1. Graphical Method
Casagrande has given a general solution to determine the value of for various degrees of inclination of
the discharge face. The inclination angle may be more than 90 o, especially in case of rock fill dam.

24
Let, if a is the slope angle of the discharge face with the horizontal is known, and then various values

of corresponding to a are given by Casagrande (Table).

Here, JF indicates the distance of the focus from the point, where base of parabola cuts downstream face.
The values of and can be obtained by Eqn and Table.

Table. Values of for various slope angles (a)

Estimation of seepage

Example:
An earth dam made of a homogeneous material has a horizontal filter and other parameters
as shown in the figure. Determine the phreatic line and the seepage quantity through the
body of the dam.

Figure 5 Section of a homogenous earth dam

.

25
 For the origin of the Cartesian co-ordinate system at the face of the filter (point F), the
equation of the parabola of the seepage line can be expressed as:

x 2  y 2 xS

At point A, x = 65m, and y = 20m. Inserting into the parabola equation, S = 3.07m. Working
out a few more points from the equation, the parabola can be easily drawn and corrected for
the curve at the upstream face of the dam, so as to get the seepage line.

x -1.51 0 10 15 25 30 40 45 55 65
2
y 0 9.06 69.26 99.36 159.56 189.66 249.86 279.96 340.16 400.36
y 0 3.01 8.32 9.97 12.63 13.77 15.81 16.73 18.44 20.01

The amount of seepage flow is

Q = kS
-6
= 5 * 10 * 3.07
-6 3
= 15.35 * 10 m /sec per meter width of dam

B. Homogeneous dam section without horizontal filter

The focus (F) of the parabola will be the lowest point of the downstream slope as shown in Figure
5-8. The base parabola BIJC will cut the downstream slope at J and extend beyond the dam toe up
to the point C i.e. the vertex of the parabola.

Figure Homogeneous dam section without filter

26
The seepage line will, however, emerge out at K, meeting the downstream face tangentially there.
The portion KF is known as discharge face and always saturated. The correction JK (say ∆a) by
which the parabola is to be shifted downward can be determined as follows:
‡ a
α in degrees
a  a
o
30 0.36
o
60 0.32
o
90 0.26
o
120 0.18
o
135 0.14
o
150 0.10
o
180 0.0

α is the angle which the discharge face makes with the horizontal. a and ∆a can be connected by
the general equation;

27
Seepage through dam

Filtration (seepage) computation through dams are carried out with the aim of:
• determining the position of the seepage depression curve
• finding the value of the gradient and velocity of filtration
• determine the filtration flow (discharge)
The seepage pattern through a dam is shown above. The pattern is the same
irrespective of the material (sand, clay, loam) of the dam, though the rate of
seepage will depend on soil type. The emergence of seepage lines on the
downstream slope tends to make the downstream slope unstable. Either the
downstream slope has to be made very flat or the seepage must be diverted
away from the downstream slope. The second alternative is favoured because it
is economical.
The diagram below shows the seepage lines for a homogeneous earthfill dam
with a horizontal gravel filter on the base of the dam at the downstream side. The
flow lines

20
enter the blanket vertically. Cassagrande (1937) has shown that the phreatic
line, which is the topmost seepage line, quite closely approximates a parabola.
The parabola

Seepage through a homogeneous dam with horizontal drainage blanket

(filter)

Insert drawing

21
intersects the water surface at A such that AB = 0.3 CB. Near the upstream
face, the phreatic line diverges from the parabola and join B perpendicularly.
The focus of the parabola is at F and the directrix passes through D. Every point
on the parabola with origin at the focus F can be given by

y 2 − y 02
x= ----------------1
2 y0
At the point A, x = d and y = H. Substituting in equation 1 yields

y0 = H 2 + d 2 − d ------------(2)
y0 , which is also the distance FD, is determined from eq. (2)
How to get the point D graphically. With A as the centre and AF as radius, draw
an arc to cut the water surface at J. From J drop a perpendicular to the base at
D. D is the directrix. Locate the mid-point of FD at E which is the vertex of the
parabola.
The phreatic line can be drawn by equation (1).
For an approximate estimate of seepage, the Darcy’s law at point G , per unit
length of dam
 dy 
q = KiA = K   y 0 -------------- (3)
 dx 
dy y0
From eq. (1) dx = --------------------(4)
2 xy0 + y 02
Since at G, x= 0, and y = y0 then

dy/dx = 1 ------------------------------(5)
Hence q = K (1) (y0) ------------------------------(6)
The flow through the section at G is the same as at any other section.
Consequently, the total seepage through the length L of the dam is given by:
Q = K(y0)(L) ------------------------------(7)

Worked example
A homogeneous earthfill dam has a top width of 30ft and a height of 100ft with a
freeboard of 10ft. The side slopes are 1V: 2.5H. It has a horizontal drainage
blanket at the base that extends from the downstream toe to a distance of 100ft.
The embankment has a permeability of 1.5 x 10-5 ft/sec. Determine the seepage
through the dam.

Solution
H = 90 ft ; CB = 90 (2.5) = 225ft ; AB = 0.3(225) = 67.5ft and CA = 157.5 ft
d = base length – blanket length – CA

22
 = 530 -100 – 157.5 = 272.5 ft.

From eq. (2) y 0 = H 2 + d 2 − d = [(90)2 + (272.5)2]1/2 -272.5 = 14.5 ft

From eq. 6 q = K (1) (y0) = (1.5 x 10-5)(14.5) (1) = 2.18 x 10-4cfs or 18.8ft3/day
per foot of dam.

Seepage line of a homogeneous dam without filter

For a homogeneous dam without filter, the focus point F of the parabola is at the
downstream toe of the dam. The base parabola cuts the downstream slope of the
dam and extends beyond the dam.
The seepage line emerges out at point C meeting the downstream face
tangentially. The portion CF of the dam is known as the discharge face and
always remains saturated.
The portion DC (Δa) and CF (a) are inter-related as follows:
 180 0 − α 
∆a = ( a + ∆a )  0

 400 
The table below gives the values of Δa/(a + Δa) for various values of α (angles
which the discharge face makes with the horizontal) as per Cassagrande
α in Δa/(a + Δa)
degree
30 0.36
60 0.32
90 0.26
120 0.18
135 0.14
150 0.10
180 0.0

Seepage rate calculation from flow net

Flow net is a network form by streamlines (flow lines and equi potential (head
lines) which are mutually perpendicular. In the network, the portion between any
two successive flow lines is called flow channel and the portion between any two
successive equi potential lines called flow field.

Then from Darcy’s law of flow through porous medium, considering unit
thickness,

∆h
∆q = k . ( b.1) --------------------a)
l

23
 equipotential
lines

Flow lines

If Δh = drop of head through a field,

h = total head causing flow
Δq = discharge passing through the field

If the total number of potential drops in the flow net = Nd

Then Δh = h/Nd ----------------------b)
h b
Putting b) into a), we obtain ∆q = k .   ------------c)
Nd  l 

h b N b
Or Total flow q = ∑ ∆q = k .  .N t = kh t .  ----------- d)
Nd  l  Nd  l 
Where Nt = total number of flow channels in the net.
When the flow field is square, then b = l and the discharge becomes
Nt
q = kh ---------------------f)
Nd

Seepage under the Dam

If the foundation material is alluvial sand or gravel, seepage could occur

underneath the dam. If the upward seepage pressure of water near the toe is
greater than the effective weight of the soil, the surface of the soil will rise at a
point of least resistance, and water and soil will start flowing away from dam.
This phenomena is known as piping and can result in the sliding of the toe or the
settling of the whole dam.

24
The submerged unit weight soil is given by
G −1
γ sub = s γw ----------(Y)
1+ e
where Gs = specific gravity of soil; e = void ratio.

For a seepage line at a gradient, I, the upward seepage force per unit volume is
iγw . When the two forces are in balance,

Gs − 1
i= -----------------(Z)
i+e
Eq. (Z) is known as the critical gradient and equal to unity. A gradient of slightly
higher than unit value will cause piping or sand particles to be in an unstable
condition known as quicksand. The actual gradient at the downstream end of the
dam is evaluated from the flow net by dividing the head difference between the
lst two potential lines by the distance between these potential lines. This should
be less than unity.
An empirical approach, the creep ratio, L/H is computed; here L is the length
along the surface of contact between the soil and the base of the structure. This
ratio is kept at 4 for gravels and 18 for sand and silt.

Measures against seepage. Seepage Control.

The following are some common measures for controlling seepage through the
dam and embankments.
1. Prevention of Seepage through Foundation
i) By providing drainage trenches
ii). By providing downstream seepage berms
iii). By providing impervious blanket layer on upstream slope
iv). By providing impervious cutoff.

2. Prevention of Seepage through Embankment (Dam)

i). By providing horizontal drainage filter

25
 ii). By providing toe filter
iii). By providing filter downstream of toe
iv). By providing downstream coarse section
v). By providing chimney drains extending upwards into the embankment

Chimney drain

Design of Filter

The design of filters should be done in such a way that all the seepage water
through the dam is effectively drained off. The filter consists of several layers.
The first layer of the filter which comes in contact with the seeping water consists
of fine sand material. Subsequent layers of filter are made of sand of increased
fineness. The last layer of the filter is made of gravels. The soil of the earth dam
and the foundation material surrounding the filter are known as the base material.
The filters of filter drains are known as reverse or inverted filter.
The U.S. Bureau of reclamation, Washington 1960 recommended the following
for materials to be used for filters;
i) Filter material should be fine and poorly graded so that the voids in
the filter are small and thus prevent base material from entering the
filter
ii) The filter material should be coarse and pervious in relation to the
base material. This aspect facilitates rapid removal of seeping
water without building up any seepage forces within the filter.
iii) The filter material should be coarser than the perforations of
openings in the drain pipes, so that filter material is not lost in the
drains. The perforations (openings in the pipes drains should be
adequate to admit all seeping water safely.
iv) The thickness of filter material should be sufficient to provide a
good distribution of all particle sizes, also throughout the filter. The
thickness should be adequate to provide safety against piping.
Terzaghi has recommended the following two requirements which should be
fulfilled by the filter:
a). The D15 size of the filter material must not be more than 4 to 5 times the D85 of
the base material. This prevents the foundation material from carrying through
the pores of the filter material.

26
b). The D15 size of the filter material should be at least 4 to 5 times the D16 of the
base material. This keeps seepage forces within the filter to permissible levels.
The above criteria can be expressed as follows:

D15 of filter < 4 to 5 < D15 of filter

D85 of base material D15 of base material.

The above criteria has been modified by USBR in “ Design of small Dams ” as:
a)
a) D15 of filter material = 5 to 40
D15 of base material

b). D15 of filter material = 5 or less

D85 of base material

c). D85 of filter materials . = 2 or more

Max opening of perforations of pipes

d). The grain size curve of the filter material should be about parallel to the curve
of the base material.

6. Stability of Earthfill dams

Stability computation for earthfil dams are done to check on the following:
2. stability of downstream slope during steady seepage
3. stability of upstream slope during drawdown
4. stability of downstream and upstream slopes during and
immediately after construction
5. Stability of foundation against shear.

Earthfill dams usually fail due to the sliding of large soil mass along a curved
surface.
The most common method used for examining the stability of earthfill dam
embankment slopes is the Swedish slip circle or (the slice) method.

Swedish slip Circle Method (Slide Method)

In this method, the failure of the embankment surface is assumed to be
cylindrical. The factor of safety against sliding, which is the ratio of average
shearing stress as determined by the Coulomb equation s = (c + σ tan φ) to the
average shearing stress determined by statics on a potential sliding surface.
For testing the stability of a slope, the centre of the possible arc is assumed. It is
necessary first to locate the centre of the Critical Circle to locate the Centre of
Critical Slope.
Fellenious has given the following method to locate the locus on which the
probable centre line may lie.

27
 P
β

WL
S
α
H
i

R 4.5 H
Q

For a homogeneous soil, the centre of critical slip centre lies on the line PQ. The
coordinates of point Q is H downwards from toe and 4.5 H horizontal away from
toe as shown above. The location of point P is done with the help of directional
angles i, α and β given in the table below.

Slope angle Slope Directional angles

(i)o
α β
11.3 5:1 25 35
18.4 3:1 25 35
26.6 2:1 25 35
33.8 1.5 : 1 26 35
45 1:1 28 37
60 0.58 : 1 29 40

First extend the surface of the fluid WL. Draw angle i from the toe of dam to cut
the surface of fluid at S. From S draw the angle β and from the toe draw angle α
to intercept angle β at the point P. Now by joining QP, we obtain the line on which
the centre of critical slip centre lies.
Now on line QP we assume point O and an arc AB is drawn with radius OA. The
earth mass above the arc is divided into a number of vertical segments known as
slices. The number of slices may be from 6 to 12. Neglecting the forces between
slices, each slice is assumed to act independently as a column of soil of unit
thickness and width. The weight of each slice is assumed to be acting at its
centre. The weight is resolved into two components N and T (ie. normal

28
component passing through the centre of the arc O, and will not cause any
driving moment on the slice. But the tangential component T causes a driving
moment of T x R). Near the toe of the dam, some of the slices may cause
resisting moment and in such cases, T is negative.

O
R O
A
R

N
α
B
W

Let thew slices be numbered 1, 2, …6 and their weights be numbered W1, W2…
W6
Now N = W cos α and T = W sin α
From Coulomb’s equation, the resisting force

FR = c x ΔL + N tan φ ……….1)
where c = unit cohesion ; ΔL = curved length of slice ;
φ = angle of internal friction of soil.

The driving moment

MD = R x ∑T …………………..2)
The resisting moment
MR = R[c∑ΔL + tan φ x ∑N
∑ΔL = sum of curved length of all components = 2πθ/360 - length of arc AB

The factor of safety against sliding

F .S =
[
M R R c ∑ ∆L + tan φ ∑ N
=
]
MD R∑ T

F .S =
[c∑ ∆L + tan φ ∑ N ] .................(3)
∑T
The values of ∑N and ∑T are generally calculated in a tabular form.

29
Now if w1, w2, …….wn are the weight of slices, then
∑N = cos α (∑w) and ∑T = sin α∑(w)

No. of slice Weight of N = w x cos α T = w x sin α c x ΔL

slice
1 w1 N1 T1 c x ΔL1
2 w2 N2 T2 c x ΔL2
3 w3 N3 T3 c x ΔL3
4 w4 N4 T4 c x ΔL4
.. .. .. .. ..
.. .. .. .. ..
n wn Nn Tn c x ΔLn
∑N ∑T c x ∑ ΔL =
c x 2πθ/360

The factor of safety is computed for the point O with equation (3). Different
points on the line QP are chosen and the procedure gone through to deyermine
the factor of safety. Various factor of safety so obtained are plotted as ordinates
on the corresponding centres and a smooth curve of F.S drawn. The centre
corresponding to the lowest factor of safety is the required centre of critical slip
circle.

Downstream Slope Stability during Steady Seepage

The critical condition for downstream slope occurs when the reservoir is full and
seepage is at its maximum rate. The seeping water below the phreatic surface
exerts a pore pressure on the soil mass. Therefore, the shearing strength of the
slices of the critical arc within the range of pore-pressure is reduced.
The net shear strength of the affected slice is

c x ΔL = (N –U) tan φ ………………(5)

where U = the pore pressure

The factor of safety (F.S) for the entire slip circle is

F.S. = c∑ΔL + tan φ (∑N - ∑U) = c∑ΔL + tan φ ∑N’ …….(6)

∑T ∑T

where N’ = normal components, calculated on the buoyant unit weight

γ’=(ρs – ρw)g of the dam.
Values of ∑T is calculated on the basis of its saturated weight.

Slope Stability during Sudden Drawdown

If dam is suddenly emptied, it may cause critical conditions for the stability of the
upstream slope. In this condition, the soil pores remain filled with water causing

30
the water level to remain the same as before the sudden drawdown. The water
weight within the soil tends to slide the upstream slope along a circular arc. The
hydrostatic force acting along the upstream slope when reservoir is full is also
removed with the sudden drawdown. Therefore, the tangential component of the
saturated weight is the main force causing disturbance to the upstream slope.
The shear resistance is considerably reduced due to development of pore
pressure on the likely slip surface.
The factor of safety can be calculated by the formula (6) above.

Stability of Upstream and Downstream Slopes during and immediately after

Construction
During construction of the dam and embankment with relatively impervious soil,
8excess pore pressure is developed in the air and water entrapped in the pore
space. This is due to the fact that the soil mass undergoes a change in volume
during compaction during and after construction. With time, this initial excess
pore pressure gets gradually dissipated.

Stability of Foundation against Shear.

At dam site, the silt and clay or fine, loose cohesionless material that form the
foundation may have good imperviousness, but are weak in shear and always
require checking.
For increasing the shear area and keeping the shear stress within permissible
limits, the embankment slopes are flattened or berms are added on both sides of
the dam.

0.4B

Toe
h1 h

h2 J L Foundation material

Firm stratum B
The method for determining the factor of safety, which is approximate is based
on the assumption that earthfill materials have an equivalent liquid weight, which
would produce the same shear stress as the material will develop itself.
The horizontal shear on left is given by:

31
 h12 − h22  φ' 
P= .γ m tan 2  45 o − ...................(7)
2  2
  φ ' 
where γ m . tan 2  45 o − .........................equivalent.liquid .unit.weight
  2 

φ’ = equivalent angle of friction

( γ .h . tan φ + c )
tan φ ' = m 1 …………………(8)
γ m .h1

where φ = angle of repose of foundation material

c = unit cohesion of foundation material
γm = mean unit weight of the dam and foundation weighted in proportion
to the
depth of each.

γ d ( h1 − h2 ) + γ f .h2
γm =
h1

where γd = density of the dam; γf = density of foundation material

The average unit shear = Sav = P/b

Hence the maximum unit shear = Smax = 1.4 x Sav

The maximum unit shear occurs at 0.4 B from point J.
The unit shear strength below toe K
S1 = c + γf . h2 tan φ
The unit shear strength at point J
S2 = c + γf . h1 tan φ
Therefore the average shear strength
S = (S1 + S2)/2
The factor of safety against shear
F.S. = S/Sav
The factor of safetyso obtained must be more than 1.5 for stability of foundation
against shear.
The F.S. at maximum shear, point L, let
S = c + γav . h . tan φ = (γd . h + γf . h2)/(h + h2)
And factor of safety (F.S.) = S/Smax
For stability the value of F.S. so obtained must be more than 1.0.

Chapter 3

GRAVITY DAMS

32
Addis Ababa University Hydraulic Structures I

H
W  3 , For very low dams 4.1
5
W  0 . 55 H  0 .2 H , For dams lower than 30m 4.2
1
W  1 . 65  H  1 . 5  3 , For dams higher than 30m 4.3
Where: H is the height of the dam.
5.5. Seepage analysis
Seepage occurs through the body of all earthen dams and also through their pervious foundation.
The phreatic surface of the seepage regime, i.e. line within the dam section below which there is
positive hydrostatic pressures in the dam, must be kept well clear of the downstream face to avoid
high pore water pressures which may promote slope instability.

The amount of seepage can be easily computed from the flow net, which consists of two sets of
curves, known as „Equipotential line‟ and „stream lines‟, mutually perpendicular to each other.
For homogeneous embankments dam, discharge per unit width (q) of the dam passing through a
flow net is described as:

N f
q  kH
Nd
Where: H is the head differential.
Nf is number of stream lines.
Nd id number of Equipotential lines.
5.6. Determination of Phreatic Lines
It is absolutely essential to determine the position of the phreatic line, as its position will enable to
determine the following:
i. The divide line between the dry (or moist) and submerged soil.
ii. The top stream line and hence, helps us in drawing the flow net.
iii. To ensure that the phreatic line doesn‟t cut the downstream face of the dam, which is
extremely necessary for preventing softening of the dam.

A. Homogeneous dam section with horizontal filter

It has been found that the seepage line is pushed down by the filter and it is very nearly parabolic
except near its junction with u/s face. Since the u/s face of the dam (i.e. GB in Figure 5-6)
becomes an equipotential line when fully covered with water, the seepage line shall be
perpendicular to the face near its junction point B.

Figure 5-6 Seepage through homogenous dam section with horizontal filter

Apr-13 70
Addis Ababa University Hydraulic Structures I

Equation of the base parabola

Let a base parabola with focus at F is drawn and produced so as to intersect the water surface at a
point A as shown in Figure 1-1.Taking the focus (F) as the origin, equation of the parabola p(x, y)
can be written as
x  y  x  FD
2 2

Where; FD is the distance of the focus from the directrix, called focal distance
and is represented by S.

Hence the equation of the parabola of the seepage line becomes:

x  y  xS
2 2

Location of A is approximately 0.33HB horizontal distance upstream from point B according to

Cassagrande. Where, H is the projection of the point G on the water surface.

If the horizontal distance between the already determined point A and the focus (F) is taken as
say b, then (b, H) represents the coordinates of the point A on the parabola. And hence;
b H bS
2 2

S  b H b
2 2

The center point (C) of FD will then be the vertex of the parabola. When x = 0, y= S. Hence the
vertical ordinate FJ at F will be equal to S.
Knowing the points A, C, and J and working out a few more points from the equation, the
parabola can be easily drawn and corrected for the curve BI, so as to get the seepage line BIJC.
The amount of seepage can also be calculated easily from the equation of the seepage line as
derived below.

Darcy‟s law is defined as, q = KiA. When steady conditions have reached, the discharge crossing
any vertical plane across the dam section (unit width) will be the same. Hence, the value i and A
can be taken for any point on the seepage line

dy
i 
dx
A  y *1
dy
q  K y
dx
But from the equation of the parabola,
y  S  2 xS
2

1 
 
1

q  K  . S  2 xs
2
  2
1
.2 S  . S
2
 2 xs
2 
q  KS

Apr-13 71
Addis Ababa University Hydraulic Structures I

Example:
An earth dam made of a homogeneous material has a horizontal filter and other parameters as
shown in the figure. Determine the phreatic line and the seepage quantity through the body of the
dam.

A
Ø = 25°
2 :1 c = 2 4 kN /sq.m
 dry = 1 8.0 kN /cu.m 3 :1
25

 sub = 1 2.0 kN /cu.m

20
k = 5 *1 0^-6 m /sec

H orizon tal F ilter

F
25 25 5 15 60
8

1 30
Ø = 12°  dry = 1 8.3 kN /cu.m
c = 5 4 kN /sq.m  sub = 1 8.3 kN /cu.m

Figure 5-7 Section of a homogenous earth dam

For the origin of the Cartesian co-ordinate system at the face of the filter (point F), the equation
of the parabola of the seepage line can be expressed as:

x  y  xS
2 2

At point A, x = 65m, and y = 20m. Inserting into the parabola equation, S = 3.07m. Working out
a few more points from the equation, the parabola can be easily drawn and corrected for the curve
at the upstream face of the dam, so as to get the seepage line.

x -1.51 0 10 15 25 30 40 45 55 65
2
y 0 9.06 69.26 99.36 159.56 189.66 249.86 279.96 340.16 400.36
y 0 3.01 8.32 9.97 12.63 13.77 15.81 16.73 18.44 20.01

20
25

P hreatic line

25 25 5 15 60
130

The amount of seepage flow is

Q = kS
= 5 * 10-6 * 3.07
= 15.35 * 10-6m3/sec per meter width of dam

Apr-13 72
Addis Ababa University Hydraulic Structures I

B. Homogeneous dam section without horizontal filter

The focus (F) of the parabola will be the lowest point of the downstream slope as shown in Figure
5-8. The base parabola BIJC will cut the downstream slope at J and extend beyond the dam toe up
to the point C i.e. the vertex of the parabola.

Figure 5-8 Homogeneous dam section without filter

The seepage line will, however, emerge out at K, meeting the downstream face tangentially there.
The portion KF is known as discharge face and always saturated. The correction JK (say ∆a) by
which the parabola is to be shifted downward can be determined as follows:
a
α‡ in degrees
a  a
o
30 0.36
60o 0.32
90o 0.26
120o 0.18
135o 0.14
150o 0.10
180o 0.0

α is the angle which the discharge face makes with the horizontal. a and ∆a can be connected by
the general equation;
 180 o   
 a   a   a  o 
 400 
Example
An earth dam made of a homogeneous material has the coefficient of permeability K= 5*10 -4 cm/
sec and the other parameters are as shown in the Figure 5-9. Determine the phreatic line and the
seepage quantity through the body of the dam.
RL = 200.0m
RL = 197.5m

2:1
H = 19.5m 3:1
RL = 178.0m

66m 4.5m 44m

Figure 5-9 Body of homogeneous earth dam

‡
α will be equal to 180o for a horizontal filter case and it will be less than 90o when no drainage is
provided.

Apr-13 73
Addis Ababa University Hydraulic Structures I

5.7. Stability analysis

Three considerations govern the design of an earth embankment.
i. Side slopes must be stable;
ii. Dimensions must be sufficient to control seepage;
iii. Base width must be long enough to distribute weight of dam over sufficient area to
prevent overstress in the foundation.
An earthen embankment usually fails because of the sliding of a large soil mass along a curved
surface.
5.7.1. Stability of side slopes of earth dam
Forms of side slope failure:
Toe failure: - most likely to occur when the slopes are relatively steep or when the soil below the
toe of the slope is strong.
Base failure: - occurs when the slopes are flat or when the soil below the toe is relatively weak.
Face or slope failure: - occurs only when there is a relatively weak zone in the upper part of the
slope or when there is a very strong stratum above the toe level.
The method used for examining the stability of slopes of earthen embankments is called the
Swedish Slip Circle Method or the Slices Method. It assumes the condition of plane strain with
failure along a cylindrical arc. The location of the centre of the possible failure arc is assumed.
The earth mass is divided into a number of vertical segments called slices as shown in Figure
5-10, O is the center and r is the radius of the possible failure.

Figure 5-10 Possible slip surface in Earth fill dam

The side slopes of earth dam will be stable if the soil mass is not dislodged from the slopes.
However, the soil mass in an earthen dam is subjected to forces which tend to cause movement or
sliding of the soil mass. These forces are known as actuating, driving or disturbing forces which
mainly consists of gravity forces. The movement or sliding of the soil mass in an earthen dam is
resisted by the resisting or stability forces which are provided mainly by the shearing strength of
the soil.
The stability of the side slope of an earthen dam is thus analyzed by assuming a surface slippage
within the soil mass and by determining the resisting and the disturbing forces acting on this
surface and the moments of these forces about the center of rotation, and then factor of safety
against sliding is calculated.

Apr-13 74
Addis Ababa University Hydraulic Structures I

The forces acting on the slices are:

1. The self weight W of the slice acting vertically downward through the center of gravity.
2. The cohesive forces acting tangentially opposite to the direction of probable slippage
3. The soil reaction across the arc. When the soil mass is about to slide, the soil reaction will
act at an angle ф (the angle of internal friction of the soil) to the normal i.e. radial
direction
4. The soil reaction on the two vertical sides of the slice exercised by the adjacent slices on
the right and left respectively.
5. Pore pressures at the base of the arc, and left and right side of the slice.
Usually it is assumed that the soil reactions on the two vertical sides of the slice cancel each other
and so also the pore pressures on the two sides balance each other
The disturbing force is the component of weight of slice in tangential direction i.e.,
T  W sin  ,
Where: α is the angle which the slope makes with the horizontal.
The total disturbing forces will be summation of disturbing forces for all slices;

T  T1  T 2  T 3  ....

The total disturbing moments over the sliding surface will be equal to
M d
 T r i i
 r  Ti

The magnitude of shear strength developed in each slice will depend upon the normal
components of that slice. Its magnitude will be:
 c  L  N tan 
Where; c is the unit cohesion of the soil
∆L is curved length of the slice
Ф is the angle of internal friction
N is equal to Wcosα
The total resisting force will be summation of resisting forces for all slices;
  c  L   N tan 
 c L   N  tan 
The total resisting moment over the entire sliding surface will be equal to
M r
 r c   L   N  tan  
Hence the factor of safety against sliding
M c   L  tan   N
FS  r

M d T
For determining the stability of the proposed side slope of an earth dam it is necessary to find the
least factor of safety which may occur on any of the possible surfaces of slippage or slip circles.
Slip circle which yields the least factor of safety is the most critical and hence it is known as
critical surface slippage or critical slip circle. For locating the critical surface of slippage, it is
necessary to try several different surfaces of slippage as one trial gives the value of factor of
safety for that arc only.

Apr-13 75
Addis Ababa University Hydraulic Structures I

For preliminary analysis 4 to 5 slices may be sufficient; however, 10 to 15 slices are considered in
general. It is not necessary for the analysis to make all the slices of equal width, but for the sake
of convenience it is customary to have slices of equal width.
In order to reduce the number of trials, Fellenius has suggested a method of drawing a line,
representing the locus of the critical slip circle. The determination of this line PQ is shown in
Figure 5-11. The point P is obtained with the help of directional angles α1 and α2 as shown in
Table 5-4

a. downstream slope b. upstream slope

Figure 5-11 locus of critical circle

Table 5-4 Slope and respective directional angle

Directional angles
Slope α1 in degrees α2 in degrees
1:1 27.5 37
2:1 25 35
3:1 25 35
4:1 25 35
5:1 25 35

Design parameters to be employed in stability analysis may be summarized as follows:

a. Stability of downstream slope during steady seepage (reservoir full)
The most critical condition for the d/s slope occurs when the reservoir is full and the seepage is
taking place at full rate.
The seepage water below the phreatic line exerts a pore pressure on the soil mass which lies
below the phreatic line, see Figure 5-12.
b

α4
V41
Phreatic line
h4 V42

N4
W4 τ4

Apr-13 76
Addis Ababa University Hydraulic Structures I

Figure 5-12 stability of Downstream slope during steady seepage

Consider slice number 4 in Figure 5-12, the weight of the slice is defined as;
W 4   dry * V 41   sat
* V 42

α4 read from the scaled drawing of the earth fill dam. And tangential component of W4 is defined
as, whish is shear stress developed at failure plane,
T 4  W 4 sin  4
The pore pressure for slice 4 is represented by the piezometric head h4. Hence pore water pressure
is
U w4   w h4

Shear strength developed for the slice is quantified from two soil parameters, apparent cohesion c
and angle of shearing resistance ф. Shear strength at failure plane is defined as
 4  cL 4  W 4 cos  4   w h 4 L 4  tan 

b
Where; L4 is
cos  4
The factor of safety of slide 4 is
T4 cL 4  W cos  4   w h 4 L 4  tan 
FS 4  
4 W 4 sin  4
The factor of safety for the entire circle is then given by the equation

FS 
c L i i
  W cos   
i i w
h i L i  tan 

 W sin 
i i

b. Stability of Upstream Slope during sudden drawdown

For the u/s slope, the critical condition can occur, when the reservoir is suddenly emptied. In such
case, the water level within the soil will remain as it was when the soil pores were full of water.
The weight of this water within the soil now tends to slide the u/s slope along a circular arc.
The tangential components of the saturated soil lying over the arc will create a disturbing force;
while the normal component minus the pore pressure shall supply the shear strength of the soil.
Table 5-5: General format of computation
Slice# W T N U l Ul N‟=N-ul tanΦ N‟tanΦ Cl N‟tanΦ+cl
1
…
n
ΣT Σ (N‟tanΦ+cl)

The factor of safety is finally obtained from the equation

c N
Li  tan 
'
i
FS 
T
'

Where: N’ represents normal components on submerged density

T’ represents tangential components on saturated unit weight of the soil

Apr-13 77
Addis Ababa University Hydraulic Structures I

5.7.2. Stability of earth dam against horizontal shear developed at the base
of the dam
Approximate method for checking the stability of u/s and d/s slopes under steady
seepage from consideration of horizontal shear at base
1. Stability of u/s slope during sudden drawdown
It is based on the simple principle that a horizontal shear force Pu is exerted by the saturated soil.
The resistance to this force Ru is provided by the shear resistance developed at the base of the soil
mass, contained within the u/s triangular shoulder GMN of Fig.
Considering unit length of the dam, the horizontal force Pu is
 h 2 2  h 
2

Pu   1 tan  45     w 1 
 2  2 2 

 sub h1   dry  h  h1 
1 
Where; h
Shear resistance Ru of u/s slope portion of the dam developed at base GN is given by
 1 
R u  cB u  W tan   cB u    sub B u h  tan 
 2 
Where; W is the weight of the u/s triangular shoulder of the dam

The factor of safety against can be easily calculated, using

Ru
FS 
Pu

It should be more than 1.5.

2. Stability of d/s slope under steady seepage
It is based on the consideration of horizontal shear at base under the d/s slope of the dam. The
horizontal shear force Pd is given by, referring Fig;
 h 2 2  h 
2

Pu   2 tan  45     w 2 
 2  2 2 

 sub h 2   dry  h  h 2 
2 
Where; h
Shear resistance Rd of d/s slope portion of the dam is given by;
R d  cB d  W tan   cB u   dry A1   sub
A 2  tan 

The downstream profile RTS of the downstream slope portion of the dam has an area A1 and of
dry soil above seepage line and the area of submerged soil say A2 below the seepage line.
The entire weight W may be calculated on the basis of submerged soil as it will be on a still safer
side. In that case,
W   sub B d h

The factor of safety against shear can be easily determined as;

Apr-13 78
Addis Ababa University Hydraulic Structures I

Rd
FS 
Pd

5.8. Foundation analysis

Foundation stress in earth dams are not usually critical except when the foundation material
consists of unconsolidated clay or silt with low shearing strength.
Consider a dam on homogeneous, unconsolidated earth foundation of thickness t

Figure 5-13 Homogenous embankment dam with pervious foundation of thickness t

The downward force exerted on the foundation at the center of the dam tends to squeeze the
foundation material from under the dam. But shear stress develops in the foundation resisting this
action. Assuming the foundation loading to vary as indicated above, Leo Jugenson suggested the
following maximum stresses:
If t > L, τmax= 0.256γf Hs ; Where γf = specific weight of fill
Usually t < L
If t < L/10, τmax = γf Hs t/L
Shear strength = Ss = c + σ tanΦ
The factor of safety against overstress is
FS = Ss/ τmax
A minimum value of FS = 1.5 is recommended.
Example:
Design the embankment dam shown in Figure 5-7 used as an exaple for analysis of seepage flow.
Detail all the necessary procedures and important consideration in the process.
Design Solution
The stability design process starts by determining the phreatic line profile which is done before.
The critical slipage circle is then drawn by following the suggestion made by Fellinus. Here a
single slipage circle is considered for illustration and four slices were considered for both
upstream and downstream slope failure. The geometric informations were then determined as
shown below.

Apr-13 79
Addis Ababa University Hydraulic Structures I

1 2 0 .2
5

A1 A1

7 1 .3
A2 A2
25

20
1 3 .8
P hreatic line

1 2 .5
A3 A3
A4 A4
25 25 5 15 60
130

Fig. Sample of failure circle, slices and related measurments of the earth dam section
Geometric properties of slices
Upstream Downstream
Slice Area(m ) 2
 (deg) hw (m) L (m) 2
Area (m )  (deg) L (m)
A1 25.97 32 0 14.5 68.58 40 18
A2 111.68 24 6.1 22 111.48 28 15.4
A3 120.83 15 6.2 20.7 96.1 17 14.2
A4 55.52 4 3.6 20.1 40.11 7 13.7

Area in relation with phreatic line.

Area (m2) Dam U/s shoulder D/s shoulder
Under seepage line (saturated) 1102.08 838.1 221
Dry portion 584.92 99.4 404
Total 1687 937.5 625
To assess the overall stability of the dam considering 1m length,
Dam U/s shoulder D/s shoulder
Item Area(m2) Weight(kN) Area(m2) Weight(kN) Area(m2) Weight(kN)
Under seepage line 1102.08 13224.96 838.1 10057.2 221 2652
Dry portion 584.92 10528.56 99.4 1789.2 404 7272
Total 1687 23753.52 937.5 11846.4 625 9924

The stability design then proceeds by first considering the entire embankment and its interaction
with the foundation.
Shear resistance of the dam at the base(R)
R = C + W tanΦ
Where: C = total cohesive resistance of the soil at the base
= c*B*1 = 24 * 130 * 1
= 3120 kN
Wtan = 23753 * tan250
= 11076 kN
R = 3120 + 11076
= 14196 kN
Horizontal force due to hydrostatic pressure of water
P = ½ w h2 = ½ * 10 * 202

Apr-13 80
Addis Ababa University Hydraulic Structures I

= 2000kN
Factor of safety against failure due to horizontal shear at the base
FS = R/P = 7.1 > 1.5 Safe!

Check stresses in the foundation

t = 8m < L/10 = 130/10 =13,
Hence,
τmax= γf Hs t/L
= 18.3 * 20* 8/130
= 22.52kN/m2
Shear strength = Ss = c + σ tanΦ
= c + W/L tan
= 54 + 23754/130 * tan 120
= 92.8 kN/m2
The factor of safety against overstress is
FS = Ss/ τmax = 92.8/22.52 = 4.12 > 1.5 Safe!

Stability of u/s and d/s slopes against sliding shear.

Upstream slope (under sudden drawdown):
Considering unit length of the dam, the horizontal force Pu is
 sub h1   dry  h  h1 
1 
h
12 * 13 . 8  18  25  13 . 8 

25
 14 . 7
 h 2 2   h1 
2

Pu   1 tan  45     w 
 2  2  2 

 14 . 7 * 25 2 2 25  13 . 8 
2

  tan  45    10 
 2  2  2 
 2816 . 6

Shear resistance Ru of upstream slope portion of the dam developed at base GN is given by,
R u  cB u  W tan   54 * 75  11846 . 4 * 0 . 47  9574 . 1
Where; W is the weight of the upstream triangular shoulder of the dam.

The factor of safety against shear can be easily calculated,

Ru 9574 . 1
FS    3 .4  1 .5 Safe!
Pu 2816 . 6

It has been known that the maximum intensity of shear stress occurs at a distance 0.6Bu (where Bu
is the base length of the upstream shoulder) from the heel and is equal to 1.4 times the average
shear intensity.
Hence, maximum shear stress induced (τmax) = 1.4(Pu/Bu)
= 1.4 (2816.6/75)
= 52.6
The unit shear resistance developed at the same point is
τf = c + 0.6hsubtan

Apr-13 81
Addis Ababa University Hydraulic Structures I

= 24 + 0.6 * 25 * 12 * tan 250

= 107.9
FS at the point of maximum shear should be greater than unity.
FS = τf / τmax = 107.9 / 52.6 = 2.0 > 1 Safe!

For the downstream shoulder, similarly,

12 * 12 . 5  18  25  12 . 5 
1   15
25
 15 * 25 2 2 25  12 . 5 
2

Pu   tan  45    10   2683 . 7
 2  2  2 
R u  cB u  W tan   54 * 60  9924 * 0 . 47  7904 . 3
Ru 7904 . 3
FS    2 .9  1 .5 Safe!
Pu 2683 . 7

Maximum shear stress induced (τmax) = 1.4(Pu/Bd)

= 1.4 (2683.7/50)
= 75.1
The unit shear resistance developed at the same point is
τf = c + 0.6hsubtan
= 24 + 0.6 * 25 * 12 * tan 250
= 107.9
FS at the point of maximum shear should be greater than unity.
FS = τf / τmax = 107.9 / 75.1 = 1.44 > 1 Safe!

Apr-13 82
Addis Ababa University Hydraulic Structures I

Analysis of upstream and downstream slopes by Swedish Circle method:

Upstream slope
Slice Area  W T N U L UL N‟=N-ul tanΦ N‟tanΦ Cl N‟tanΦ+cl
A1 25.97 32 467.46 247.72 396.43 0 14.5 0 396.43 0.47 186.32 348 534.32
A2 111.68 24 1340.16 545.09 1224.3 61 22 1342 -117.7 0.47 -55.32 528 472.68
A3 120.83 15 1449.96 375.28 1400.55 62 20.7 1283.4 117.15 0.47 55.06 496.8 551.861
A4 55.52 4 666.24 46.47 664.62 36 20.1 723.6 -58.98 0.47 -27.72 482.4 454.68
ΣT 1214.56 Σ (N‟tanΦ+cl) 2013.54

c Li  N tan 
'
i 2013 . 54
FS    1 . 66  1 . 5 Safe!
T
'
1214 . 56

Downstream slope
Slice Area  W T N l N‟=N-ul tanΦ N‟tanΦ Cl N‟tanΦ+cl
A1 68.58 40 1234.44 793.48 945.64 18 945.64 0.47 444.45 432 876.45
A2 111.48 28 2006.64 942.06 1771.76 15.4 1771.76 0.47 832.73 369.6 1202.33
A3 96.1 17 1729.8 505.74 1654.22 14.2 1654.22 0.47 777.48 340.8 1118.28
A4 40.11 7 721.98 87.99 716.6 13.7 716.6 0.47 336.8 328.8 665.6
2329.27 Σ (N‟tanΦ+cl) 3862.66

3862 . 66
FS   1 . 66  1 . 5 Safe!
2329 . 27

Apr-13 83
Addis Ababa University Hydraulic Structures I

5.9. Internal drainage system

General
Purpose of drainage:
1. To reduce the pore pressure thereby increasing the stability of the dam
2. To prevent piping so that soil particles are not carried away from the embankment.
A drainage system consists of two components.
a. the protective filter
b. the conduit which collects and disposes of the seepage.
Protective filter: - serves to allow free drainage and to prevent erosion.
It is provided between
Riprap and embankment
Core and embankment
Embankment and drains
Experiments by Terzaghi, Bertram and others have shown that a filter need only hold the coarse
15% of the grain size. These coarse particles D85 and over, will collect over the filter opening
bridging over it and trapping finer particles.
Size of filter holes, Df  D85 ( of the soil being filtered)
From tests, the following criteria are established.
D15 (filter)  4 to 5 D85 (soil); to satisfy prevention of migration of soil particles.
D15 (filter)  4 to 5 D15 (soil); for free drainage.
Filter gradation curve should be smooth and parallel to the soil being filtered. When the soil is
gap graded recompute and re-plot the grain size distribution using only the fraction finer than the
break as representing the entire soil; apply the filter criteria to this distribution
Filter thickness:
Thin filter is desirable to minimize flow resistance. Practical considerations, however, put
minimum sizes as shown in Table 5-6.
Table 5-6: Representative mean filter thickness
Filter Thickness for given head, cm
0 – 25 m 25 – 50 m 50 – 100 m
Fine sand 15 30 45
Coarse sand 25.5 45 60
Gravel 30 60 75
For every fine grained soil a multiple layered filter is necessary. Each successive layer is designed
to fit the D15 and D85 of the finer layer it must filter. The last layer must fit the openings of the
drain conduit which carries the water away.

Apr-13 84
Addis Ababa University Hydraulic Structures I

Example:
Determine the size of the bed material for the embankment dam with the grain size shown below.
Grain size Riprap Dam
D15 70 mm 0.3 mm
D85 130 mm 2.0 mm

Apr-13 85
Addis Ababa University Hydraulic Structures I

Drain Conduit:
Function: to collect water from the filter and carry it away with as little head as possible.
Quantity of Flow: Estimated from the seepage analysis. A factor of safety of 5 is not uncommon.
Simplest conduit: uniform coarse fragmental material (coarse sand, gravel, crushed rock, etc.)
A properly designed filter must surround the drain. For high discharges or when suitable crushed
rock is not available, pipe conduits wig perforated and flexible joints are employed.
Types of drain
Trench drain:
Trench drain is used for intercepting seepage through homogeneous foundations and those
containing horizontal pervious strata or seams

Figure 5-14 Trench Drain

Mound drain: used when the need for embankement drainage exceeds that of the foundation.

Figure 5-15 Mound drain

Position- when there is cut-off, the drain is placed immediately downstream from the cut-off to
relieve any pressure build up.
When there is no cut-off, the position depends on the seepage analysis of dam and foundation.
Upstream location
Increases stability at the cost of increased seepage and cost of conduit.
Minimum distance from downstream toe to provide substantial increase in stability is 1/3 of the
base width. Maximum distance is 2/3 of base width, from downstream toe.
Blanket drain:
This is horizontal drain placed on top of foundation.
To intercept water from vertical fissures in the foundation;
To lower the seepage line in the embankment

Apr-13 86
 Addis Ababa University Hydraulic Structures I

Figure 5-16 Blanket drain

Riprap:
Riprap is required on the upstream slope and the downstream slope below the tail water level. An
estimation of the required weight of rock pieces required for riprap is given by Hudson as
W  (H2 ρst tan α) / (3.2Δ2)
Where: the factor 3.2 is for smooth quarry stone
ρst = density of rock
α = angle of slope
Δ = (ρst – ρw)/ ρw
ρw = density of water
Types of riprap
Dumped riprap: consists of angular broken rock dumped from truck and spread.
Hand-placed riprap: consists of more or less prismatic stone placed on end to form rough
pavements.

Figure 5-17 Types of riprap

Apr-13 87

Powered by TCPDF (www.tcpdf.org)